Byron/gitoxide git-discover-v0.9.0

git-discover v0.9.0

on GitHub

Bug Fixes

• Discover repo with relative path and ceiling
  A couple of problems are repaired to allow discovering a repository from
  "." with a ceiling directory.

  One problem was that find_ceiling_height() did the wrong thing when
  confronted with any relative search_dir. This is resolved by converting
  search_dir to be absolute if it is relative.

  The other problem was that discover_opts() also mishandled relative paths.
  When the cursor started out as ".", cursor.pop() would be blindly called
  such that cursor would be "" for the second iteration. When a ceiling
  directory was in use such that there was a max_height, the current height
  would be burned going from ".", to "", and then to "", before finally
  actually getting to a real parent directory. This problem is ameliorated by
  testing whether the cursor has a non-empty parent before popping.

  N.B. the new test case relies on the test running from the git-discover
  directory such that the gitoxide repository will be found. This is a bit
  fragile and will fail if, for example, the test is run from an unpacked
  gitoxide source tarball.

New Features (BREAKING)

• Path::from_dot_git_dir() now takes the current_dir as argument and returns Option<path>
  That way it's possible to avoid at least one call of
  std::env::current_dir() per invocation, which also is more consnstent
  with similar plumbing methods.

  Furthermore it can signal with None if the input directory was invalid.

• upgrade edition to 2021 in most crates.
  MSRV for this is 1.56, and we are now at 1.60 so should be compatible.
  This isn't more than a patch release as it should break nobody
  who is adhering to the MSRV, but let's be careful and mark it
  breaking.

  Note that git-features and git-pack are still on edition 2018
  as they make use of a workaround to support (safe) mutable access
  to non-overlapping entries in a slice which doesn't work anymore
  in edition 2021.

Commit Statistics

• 9 commits contributed to the release over the course of 2 calendar days.
• 4 days passed between releases.
• 3 commits were understood as conventional.
• 0 issues like '(#ID)' were seen in commit messages

Commit Details